Background: A laparoscopic approach to pyloromyotomy for infantile pyloric stenosis has gained popularity but its effectiveness remains unproven. We aimed to compare outcomes after open or laparoscopic pyloromyotomy for the treatment of pyloric stenosis. Methods: We did a multicentre international, double-blind, randomised, controlled trial between June, 2004, and May, 2007, across six tertiary paediatric surgical centres. 180 infants were randomly assigned to open (n=93) or laparoscopic pyloromyotomy (n=87) with minimisation for age, weight, gestational age at birth, bicarbonate at initial presentation, feeding type, preoperative duration of symptoms, and trial centre. Infants with a diagnosis of pyloric stenosis were eligible. Primary outcomes were time to achieve full enteral feed and duration of postoperative recovery. We aimed to recruit 200 infants (100 per group); however, the data monitoring and ethics committee recommended halting the trial before full recruitment because of significant treatment benefit in one group at interim analysis. Participants, parents, and nursing staff were unaware of treatment. Data were analysed on an intention-to-treat basis with regression analysis. The trial is registered with ClinicalTrials.gov, number NCT00144924. Findings: Time to achieve full enteral feeding in the open pyloromyotomy group was (median [IQR]) 23·9 h (16·0-41·0) versus 18·5 h (12·3-24·0; p=0·002) in the laparoscopic group; postoperative length of stay was 43·8 h (25·3-55·6) versus 33·6 h (22·9-48·1; p=0·027). Postoperative vomiting, and intra-operative and postoperative complications were similar between the two groups. Interpretation: Both open and laparoscopic pyloromyotomy are safe procedures for the management of pyloric stenosis. However, laparoscopy has advantages over open pyloromyotomy, and we recommend its use in centres with suitable laparoscopic experience. Funding: Sir Arthur Halley Stewart Trust.